Omada Health Delivers Machine Learning-Enhanced Preventative Virtual Care Using AWS

In 2021, Omada Health unveiled the Omada Insights Lab, an internal data analysis engine that leverages over one billion proprietary data points from over half a million members’ real-world interactions with Omada Health programs to yield recommendations for optimizing and personalizing care. The Insights Lab uses Amazon Redshift, a cloud data warehouse, to securely analyze health data with machine learning models.

“AWS empowers us to digest an enormous amount of data and process it quickly to provide actionable, metrics-based alerts for clinicians,” said Jasik. “We couldn’t generate timely alerts if we didn’t have a scalable, compliant, and cost-effective platform that had these capabilities.” For example, AWS tools enable data ingestion from continuous glucose monitors that produce 96 data points per day per patient, yielding deeper insights that can improve care and health outcomes for prediabetic and diabetic program members on both individual and population-wide levels.

By migrating compute architecture from Ruby on Rails to AWS Lambda, a serverless, event-driven compute service that runs code without provisioning or managing servers, Omada Health was able to reduce overall computing costs by 30 percent. “AWS Lambda has enabled us not only to minimize costs but also increase throughput—we can run far more insights in a shorter amount of time than we could with a Rails application,” said Myles Guanzon, vice president of engineering at Omada Health. “Since we are a cloud-native, virtual-first care provider, finding ways to pare down computing costs also enables us to reinvest those savings into different capabilities.”

Omada Health platforms are designed for program members to interact seamlessly with healthcare providers who understand their needs. To communicate with program members and provide reliable access, the Omada Health platform uses Amazon Simple Notification Service (Amazon SNS) and Amazon Simple Queue Service (Amazon SQS), fully managed messaging services for both application-to-application (A2A) and application-to-person (A2P) communication. As a fully cloud-based healthcare provider, Omada Health prioritizes security and compliance alongside ease of use and personalized algorithm development and outreach.

“It has become extremely important that companies and virtual-first care providers meet the moment with evidence-based solutions and solutions that are clinical grade,” said Jasik. “Our solutions enable healthcare to meet people where they need it, when they need it, and give program members another tool to support them in their health and wellness journey.”

By using algorithms to inform clinical decisions and then feeding the resulting data back into Omada Health’s Insights Lab to enable further optimization through machine learning, the organization prompts human care providers to interact with program members in the ways that will be most successful. As the company continues expanding to serve more individuals and cover additional disease areas, its AWS architecture will enable scaling and adaptation to incorporate insights as they emerge.

“On a global level, our relationship with AWS has been very collaborative and fruitful,” said Guanzon. “As new technologies, services, standards, and best practices get developed, we’ve been able to work with AWS to educate our teams and implement new advances that will benefit program members for better health outcomes.”

See how AWS is supporting other healthcare companies to make virtual care a reality.